Dynamic Covariance Models for Multivariate Financial Time Series
This addresses the need for accurate covariance prediction in financial modeling, offering a solution to key bottlenecks like overfitting and computational inefficiency, though it appears incremental as it builds on existing dynamic modeling approaches.
The paper tackled the problem of predicting time-changing covariances in multivariate financial time series, which suffers from overfitting, failure to capture market shifts, and high computational costs, and introduced a novel dynamic model using Bayesian inference, diffusion processes, and particle filters, resulting in excellent performance compared to standard models.
The accurate prediction of time-changing covariances is an important problem in the modeling of multivariate financial data. However, some of the most popular models suffer from a) overfitting problems and multiple local optima, b) failure to capture shifts in market conditions and c) large computational costs. To address these problems we introduce a novel dynamic model for time-changing covariances. Over-fitting and local optima are avoided by following a Bayesian approach instead of computing point estimates. Changes in market conditions are captured by assuming a diffusion process in parameter values, and finally computationally efficient and scalable inference is performed using particle filters. Experiments with financial data show excellent performance of the proposed method with respect to current standard models.